arXiv cs.CY· Pantelitsa Leonidou, Nikos Salamanos, Sotiris Gypsiotis, Michael Sirivianos·· 3 天前AI 评分20
DeTox-Fed：用联邦图神经网络检测 Fediverse 中的有毒对话
DeTox-Fed: Detecting Toxic Conversations in the Fediverse with Federated Graph Neural Networks
AI 导读
DeTox-Fed 是一个联邦图学习框架，让 Pleroma、Mastodon 等去中心化社交网络实例无需共享原始对话或审核标签，即可协作训练有毒对话分类器。各实例构建本地对话图，节点为对话树、边表示跨对话的共同用户参与，再用 GNN 在联邦学习下训练，并融合对话结构、用户互动模式、对话级统计与聚合情感信号。
正文 · 原文
Abstract:The rise of decentralized social networks (DSNs), and in particular the rapid uptake of the Fediverse (e.g., Pleroma, Mastodon, Lemmygrad), introduces new challenges in content moderation. Independent instances host their own data, follow different moderation policies, and often observe only partial views of conversations. We present DeTox-Fed, a federated graph-learning framework for detecting toxic conversations in DSNs without requiring instances to share raw conversations or moderation labels. Each instance constructs a local conversation graph, where nodes represent conversation trees and edges capture shared user participation across conversations. A Graph Neural Network is then trained in a federated learning setup, allowing instances to collaboratively learn a toxicity classifier while preserving data locality. Unlike text-only moderation approaches, DeTox-Fed combines conversational structure, user-interaction patterns, conversation-level statistics, and aggregate sentiment signals. We evaluate the framework on a large Pleroma conversation dataset and a second one from Mastodon, showing that it achieves stable toxic conversation detection under limited local labels, partial client participation, and varying moderation thresholds. Our results indicate that federated graph-based moderation is a promising direction for semi-automated moderation in decentralized social networks.
| Comments: | This revised version substantially extends the experimental evaluation by incorporating a Mastodon dataset, additional LLM models (both lightweight and top-tier), and including GraphNLI as an experimental baseline |
| Subjects: | Social and Information Networks (cs.SI); Computers and Society (cs.CY) |
| Cite as: | arXiv:2605.21054 [cs.SI] |
| (or arXiv:2605.21054v2 [cs.SI] for this version) | |
| https://doi.org/10.48550/arXiv.2605.21054 arXiv-issued DOI via DataCite |
Submission history
From: Pantelitsa Leonidou [view email]
[v1]
Wed, 20 May 2026 11:41:27 UTC (449 KB)
[v2]
Wed, 14 Oct 2026 08:29:43 UTC (269 KB)
来源：arXiv cs.CY · arxiv.org